[Users] Host IO delay high - Beancounters failure

[Henry Spanka]

Hey,
Well. I forked and developed Proxmox further by myself with ploop
integration and many other features.
However from my perspective the kernel is just the plain openvz kernel with
some drivers included.
What I'm curious about, if anybody else experienced such problems or knows
where the issue is.

The nodes are running smoothly it's just that sometimes a container freaks
out. I first had the idea,
that the SWAP is not speed limited and they're writing immense amounts of
data into the swap.
However after turning of swap, the problem still exists. Load of the host is
>1.000.

When just setting  the ram and swap attribute are you sure, the container is
limited in all cases using vSWAP?
Because many other parameters like numproc are set to unlimited. Do you guys
see any misconfiguration in the
UBCs or isolation issues?

OOM-killer tries to kill a process inside of that container. After killing
the container or waiting 5 minutes(seems like oom-killer needs
So much time), everything is normal. From my perspective the resource
limitations of OpenVZ do not apply in some cases.

----------------------------------------------------------------------------
-------------
If you have any further questions, please let us know.

Mit freundlichen Grüßen / With best regards 
Henry Spanka | myVirtualserver Development Team